[BugFix] When the value of "temperature" is 0, adjust it to 1e-06 (#4919)

* 【DataProcessor】add options thinking_mode (#4735)

* add thinking_mode

* add thinking_mode

* add thinking_mode

* add thinking_mode

* add thinking_mode

* add thinking_mode

* add unit test

* [BugFix] When the value of "temperature" is 0, adjust it to 1e-06 (#4900)

* add default temperature value

* add unit test

* update

* update

* add unit test

* update

* fix unit test

fastdeploy/engine/engine.py

@@ -251,6 +251,8 @@ class LLMEngine:
         request = Request.from_dict(task)
         llm_logger.info(f"Receive request {request}")
         if sampling_params is not None:
+            if sampling_params.temperature is not None and abs(sampling_params.temperature) < 1e-06:
+                sampling_params.temperature = 1e-06
             request.sampling_params = sampling_params
         request.preprocess_start_time = time.time()
         chat_template_kwargs = kwargs.get("chat_template_kwargs") or {}

fastdeploy/entrypoints/engine_client.py

@@ -304,7 +304,8 @@ class EngineClient:
                 api_server_logger.warning(
                     f"req_id: {data['request_id']}, reasoning_max_tokens exceeds max_tokens, the value of reasoning_max_tokens will be adjusted to match that of max_tokens"
                 )
-
+        if data.get("temperature") is not None and abs(data["temperature"]) < 1e-6:
+            data["temperature"] = 1e-6
         # logprobs
         logprobs = data.get("logprobs")
         top_logprobs = None